Neuroleptics represent the mainstay of the pharmacologic treatment of schizophrenia; however, the large number of treatment-resistant schizophrenic patients has prompted efforts at augmenting their therapeutic effects. In this study we will evaluate the clinical efficacy of alprazolam augmentation of neuroleptic treatment in treatment-resistant schizophrenic patients. Fifty-two treatment-resistant schizophrenic patients will be studied as inpatients for six weeks. During the first two weeks, patients will be stabilized on fluphenazine following which either alprazolan or placebo will be added in a double-blind manner for four weeks. Clinical response will be assessed by weekly double-blind ratings of "positive" and "negative" psychotic symptoms. In addition, plasma levels of homovanillic acid will be obtained three times weekly for assessment of alprazolan effects on plasma measures of dopamine turnover. All patients will additionally receive magnetic resonance imaging brain scans to assess the possible relationship of estimates of prefrontal cortex atrophy to favorable clinical response to alprazolam (as suggested by pilot data). This study may lead to the development of novel therapeutic approaches to the treatment-resistant schizophrenic patient as well as to a delineation of biological predictors or concomitants of favorable clinical response.